MOTIVATION Collateral blood flow plays a central role during acute ischemic stroke, and the presence of good collaterals has been linked to improved clinical and radiological outcome measures both with and without treatment. Despite this, the only method to assess adequacy of collaterals currently is invasive and expensive, digital subtraction cerebral angiography (DSA). It would be of tremendous benefit if collaterals could be visualized using a tomographic, non-invasive imaging based modality, so that early treatment decisions could incorporate this critical piece of information. AIMS AND METHODS The overall goal of this new submission is to identify and quantify cerebral blood flow (CBF) delivered via collateral routes using arterial spin labeling (ASL), a non-contrast MRI technique that combines aspects of perfusion and angiography. DSA and xenon CT (xeCT) will act as gold standards for collaterals and CBF, respectively. We aim to show equivalence between DSA and both qualitative and quantitative ASL methods optimized to visualize collaterals in 60 patients with Moyamoya disease. These patients make up an ideal population for this study, as they tend to be young and otherwise healthy, but typically have extensive collateral circulation. We also aim to demonstrate that ASL CBF measurements sensitive to slow flow, such as long post-label delay pseudocon- tinuous ASL and velocity-selective ASL, correlate with xeCT-based CBF in regions supplied by collaterals better than standard ASL and perfusion-weight- ed imaging (PWI)-based CBF. Finally, we will apply the knowledge gained from Moyamoya patients to the much larger and more relevant clinical population of acute ischemic stroke. We will obtain ASL, diffusion-weighted imaging (DWI), and bolus PWI in 60 patients both <24 hours and 3-5 days following acute stroke to determine the ability of ASL collateral CBF imaging to identify tissue at-risk of infarction, using receiver-operator characteristic (ROC) methods. SIGNIFICANCE We believe successful attainment of these aims promises to markedly improve acute stroke care by validating a non-invasive MRI-based method to assess both the presence of collaterals and the amount of CBF they deliver. Our study will lead to better understanding of ASL CBF measurements in patients with severe cerebrovascular disease and greatly enhance the already significant diagnostic power of MRI in acute ischemic stroke. PUBLIC HEALTH RELEVANCE: In the early hours following large vessel occlusion, the ultimate severity of the stroke is largely determined by the ability of collateral flow networks to supply blood to ischemic tissue via circuitous routes that bypass the proximal clot. Ro- bust collateral flow can improve response to thrombolytic therapy and decrease the risk of intracranial hemorrhage. Despite their central importance, collaterals during acute stroke are poorly understood, largely because assessment has required an invasive imaging test, cerebral angiography. This proposal assesses whether a noncontrast MRI perfusion technique, called arterial spin labeling (ASL), can yield important information about collateral flow. ASL images blend features of angiography (dependence on arterial arrival times) and perfusion (the ability to quantify CBF in absolute physiologic units). Recent technical advances and the more widespread introduction of 3T clinical MRI scanners are beginning to bring this technique into the mainstream, as a valuable and robust clinical imaging tool for the evaluation of cerebrovascular disease. To fully characterize the relationship between collateral flow and ASL, we will first study patients with Moyamoya disease. This is a chronic vasculopathy primarily of young people, who are stable enough to undergo MRI in a non-acute setting and who routinely receive cerebral angiography in preparation for possible extracranial-to-intracranial bypass. They make up an ideal population to study extensive collateral flow without the time constraints that surround acute stroke patients. We will then take the information from these studies, and apply the best ASL methods to acute stroke patients. For both groups of patients, we will validate the ASL cerebral blood flow measurements using a gold-standard, stable xenon-enhanced CT. Successful completion of this project will lead to better understanding of ASL CBF measurements in patients with severe cerebrovascular disease and greatly enhance the already significant diagnostic power of MRI in acute ischemic stroke.